Inspiration
N-TV is one of the giant news channels in Germany, they are launching their own redesigned smartphone app in a span of two weeks from now. Their focus is to create a new experience for the user and personalize it without intensifying filter bubbles. At the R U Ready summit, we carried out some user interviews to validate the idea. We talked to 6 media professionals and students about their consumption of daily news. The insight we got from the user was that most of them needed an interactive and gamified way of consuming their news articles, hence we came up with the idea of "Tinder for News".
What it does
We take into account the historic data of the user interaction with the article, where the data consists of two parameters:
- If the user VIEWED the article.
- If the user LIKED the article.
Based on these parameters we trained a recommendation system, which is based on Collaborative Filtering. We were able to find the dummy dataset to train and validate the model from here on Kaggle.
How we built it
We used a simple Lean Canvas approach and distributed the tasks for 3 Developers and 2 Media Professionals in the three lists i.e TO-DO, DOING and DONE. This way we all had a good bifurcation of tasks and we knew what each person was working on all the time. Initially, during the ideation session, we built multiple versions of paper prototypes and started improving upon it. At any point in time if anyone was stuck with a task for too long the entire team or the member would step in to help the person out.
Challenges we ran into
We ran into issues with setting up the training pipeline for our recommendation system. Designing the architecture was an iterative process. We failed a couple of times with setting up the architecture as the team is not well versed with the frontend development. But thanks to Krupesh Halvadiya who took ownership of it we managed to sail through it.
Accomplishments that we're proud of
We managed to develop a prototype on UXPin to showcase our integration with the n-tv app and also were able to make a quick working prototype to showcase the idea with a web app.
What we learned
We were able to get a good understanding of our current skillsets. Working with the media professionals in our team helped us really getting a different perspective on things. Overall it was a wholesome experience for us.
What's next for RUready Hackathon 2019 (n-tv)
We will be doing the pitch at the R U Ready Summit and see how the solution is perceived by the jury, media professionals and other visitors at the expo.
Log in or sign up for Devpost to join the conversation.